I suggest that you utilize r8Brain using ‘Power of Two’ strategy to up-sample all PCM to 705.6kHz and 768kHz and to decimate DSD to PCM, so to offload processing overhead from the FPGA architecture and WTA processing… There is a reason M Scaler exists…
I don’t see an advantage to apply DSP after up-sampling… I actually believe it is limiting relative to sample-rate support.
Note: I understand that HLC employs the r8Brain modulator, but my choice would be to have the sample-rate conversion done independent of the plug-in module.
The signal path here is 1FS to 8FS input > 16FS WTA1 filter (49,152 taps) > 256FS WTA 2 filter > 3rd order 2048 FS filter > pulse array noise shaper at 104 MHz > analogue…
If you send a 768kHz signal you skip the WTA1 filter… Do you have any empirical data that shows r8Brain is better than the WTA1? Believe me I have played with several software upscalers into the multi mega taps range… And have yet to find much benefit…
I also think you do not fully understand the use of r8Brain in the HLC… It is only used to sample rate match the FIR filters to the incoming signal sample rate… If you supply the HLC with multiple FIR filters, one for each incoming signal sample rate, r8Brain is not engaged…
Okay, then offload some processing overhead by up-sampling in Audirvāna to 352.8 or 384kHz and let WTA do it’s thing…
Do you make assessments by the numbers or with your subjective interpretation of what you hear? I absolutely realize that you are the final arbiter of the resulting sound-quality of the audition.
Hi @Antoine , this is getting a little ahead of the game, I understand, but I wanted to mention something about the remote design. (And it may actually be easier to implement before the new remote is introduced than to go back and make changes afterward.)
When looking at the Artists list in the remote, there is a helpful alphabetical index down the right hand side. This is great when I want to jump to, for example, Warren Zevon or Yevgeny Sudbin.
I like using the Folders view because that reflects my own organization of the music, even more so than an alphabetical list of artists. I have a folder for all albums that consist entirely of Beethoven works, no matter the artists/orchestra/conductor, because that is where it is easiest for me to remember to look for Beethoven’s violin quartets, for example, no matter the quartet that is playing them. However, the folder view does not have the alphabetical index running down the right hand side, meaning that I have to do a lot of scrolling to reach Zevon’s or Sudbin’s individual artist folders. I’d really like the addition of the alphabetical index to the folders view in the remote if that is possible.
Yes that is true of most DAC platforms today… I look at it from the point-of-view that is the premise for the implementation of WTA processing in the context of digital-audio performance and the reduction of noise related intrinsic jitter through the reduction of computational operations, related to bit-depth and sample-rate… Would you say that WTA is superfluous in the playback of DXD (352.8kHz) files? Don’t you think that the processing noise is reduced in the playback of a DXD sample-rate file as compared to a 16/44.1kHz file playback?
Does anyone know why I get this looping of the last second or two of every track when using convolution? It happens with every server and every DAC. Other than that, this beta works well for me.
After following discussions about the beta version for a while, I decided to try the beta today. One of the first things that caught my attention was that Audirvana EQ is designed as an embedded window.
As far as I can tell, the graph window’s aspect ratio are messed up and it has become difficult to read and control.
It would be useful if Audirvana EQ could be used as a windowed application, like other VST plugins.
I personally have no problem viewing all aspects of the Studio EQ GUI…
If you provide the information requested in the first post, along with a screenshot of what you are experiencing, this will provide more insight into the circumstances and rationale of your suggestion.
In 2019 Amirm with Audio Science Review testing found literally no jitter… And I am not much of a fan of DSD/DXD or MQA so I don’t have an opinion other than I don’t care for them…
We will need more information about your computer system resources in concert with your playback system design and Audirvāna configuration(s)… As you can see, we have some anecdotal evidence describing unexpected behaviors, however, we are not seeing best-practice approaches in implementing convolution files in the context of Audirvāna for any given user scenario… What those that are familiar with the application of computational intensive processing, is that the available computer platform resources in the context of Audirvāna playback will have great influence on the expected behavior and sound-quality.
In general terms, how were the FIR or IIR filters created? Is it specifically made for your listening room or for some place else? It may be that there is some kind of post ringing or echoing? Maybe a reverb IR?
This is a myopic snap-shot of objective evidence, measured under specific conditions, that are not related to or inclusive of the system handling of various bit-depth and sample-rate input…
So we can assume that you believe that a 16/44.1kHz recording is equivalent to a 24/88.2kHz, 24/176.4kHz or 24/352.8kHz recording?
It happens regardless. I have tried running on Dell XPS 15 (Windows 11), Mac Mini (late 2014), and NucBox G5 (Windows 11). I have tried my Chord Mojo 2 and Meridian Prime. I have tried UPnP and direct USB connection. I have tried WASAPI, ASIO, and Kernel Streaming. I have tried buffer from 1000MB to 16,000MB (32GB RAM). I have convolution filters installed for 44.1,48, 88.2, 96, 196.4, and 192. It happens on every resolution with or without upsampling and with or without EQ. It happens streaming from Tidal or Qobuz.
I would think it’s a bug in Audirvana, but why isn’t everyone seeing this? It makes no sense to me.
Your debug report may shed some light on a potential… I hope more folks chime-in on their successes and failures… You have provided a lot of insight here, and providing your debug information should top-it-off… I am as curious as you are… It would be nice to have some examples of a working scenario.
I wonder if it’s an issue with the RoPieee on Raspberry Pi4 UPnP implementation? That is a common denominator in all of my various trials.
2026-01-18 15:50:38.797 [debug]: UPnP received audio data request from 26214356 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.797 [debug]: UPnP sending audio from 2:28 playing 0:0 loaded to 3:56 from 26214356 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.823 [debug]: UPnP received audio data request from 27262932 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.823 [debug]: UPnP sending audio from 2:34 playing 0:0 loaded to 3:56 from 27262932 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.848 [debug]: UPnP received audio data request from 28311508 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.848 [debug]: UPnP sending audio from 2:40 playing 0:0 loaded to 3:56 from 28311508 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.912 [debug]: UPnP received audio data request from 29360084 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.912 [debug]: UPnP sending audio from 2:46 playing 0:0 loaded to 3:56 from 29360084 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.933 [debug]: UPnP received audio data request from 30408660 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.933 [debug]: UPnP sending audio from 2:52 playing 0:0 loaded to 3:56 from 30408660 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.955 [debug]: UPnP received audio data request from 31457236 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.955 [debug]: UPnP sending audio from 2:58 playing 0:0 loaded to 3:56 from 31457236 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.978 [debug]: UPnP received audio data request from 32505812 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:38.978 [debug]: UPnP sending audio from 3:4 playing 0:0 loaded to 3:56 from 32505812 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.000 [debug]: UPnP received audio data request from 33554388 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.000 [debug]: UPnP sending audio from 3:10 playing 0:0 loaded to 3:56 from 33554388 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.023 [debug]: UPnP received audio data request from 34602964 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.023 [debug]: UPnP sending audio from 3:16 playing 0:0 loaded to 3:56 from 34602964 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.045 [debug]: UPnP received audio data request from 35651540 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.045 [debug]: UPnP sending audio from 3:22 playing 0:0 loaded to 3:56 from 35651540 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.066 [debug]: UPnP received audio data request from 36700116 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.066 [debug]: UPnP sending audio from 3:28 playing 0:0 loaded to 3:56 from 36700116 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.089 [debug]: UPnP received audio data request from 37748692 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.089 [debug]: UPnP sending audio from 3:33 playing 0:0 loaded to 3:56 from 37748692 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.112 [debug]: UPnP received audio data request from 38797268 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.112 [debug]: UPnP sending audio from 3:39 playing 0:0 loaded to 3:56 from 38797268 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.134 [debug]: UPnP received audio data request from 39845844 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.134 [debug]: UPnP sending audio from 3:45 playing 0:0 loaded to 3:56 from 39845844 for 1048576 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:39.153 [debug]: UPnP received audio data request from 40894420 for 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:40.154 [debug]: UPnP sending audio from 3:51 playing 0:0 loaded to 3:56 from 40894420 for 873468 requested 1048576 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:50:40.154 [debug]: UPnP received audio data request from 41767888 for 307040 loaded 41767888 from buffer 0 trackId 9
2026-01-18 15:52:06.557 [debug]: UPnP Received AVTransportURI change to http://192.168.1.96:49152/audirvana/audio_f_9.wav
2026-01-18 15:52:06.557 [debug]: UPnP Buffer played, switching to 0
2026-01-18 15:52:06.674 [debug]: UPnP setPlayingBuffer discarded
2026-01-18 15:52:08.610 [debug]: UPnP send setNextAVTransportURI for track The Wolves (Act I and II)
2026-01-18 15:52:08.760 [debug]: UPnP sending audio from 3:56 playing 0:1 loaded to 3:58 from 41767888 for 307040 requested 307040 loaded 42074928 from buffer 0 trackId 9
2026-01-18 15:52:09.148 [debug]: UPnP Success setting NextAVTransportURI to http://192.168.1.96:49152/audirvana/audio_f_10.wav
2026-01-18 15:52:10.847 [debug]: FLAC decoder: HTTP reader returned less bytes 3106 than requested 0
2026-01-18 15:52:21.281 [debug]: UPnP Success sending Pause request
2026-01-18 15:52:21.285 [debug]: UPnP Received updated transport actions:Next,Previous,Play,Stop,Seek
2026-01-18 15:52:21.296 [debug]: UPnP Received new transport state PAUSED_PLAYBACK
2026-01-18 15:52:21.482 [debug]: UPnP Exiting playing start polling loop with 49 attempts remaining
2026-01-18 15:52:28.231 [debug]: UPnP Received AVTransportURI change to
2026-01-18 15:52:28.231 [debug]: UPnP Received updated transport actions:Next,Previous,Play
2026-01-18 15:52:28.262 [debug]: UPnP Received new transport state STOPPED
2026-01-18 15:52:28.336 [error]: UPnP Error 501 setting NextAVTransportURI to on instance 0
2026-01-18 15:52:28.336 [warning]: UPnP Error resetting NextAVTransportURI when stopping
2026-01-18 15:52:28.346 [error]: UPnP Error 501 setting AVTransportURI to on instance 0
2026-01-18 15:52:28.346 [warning]: UPnP Error resetting AVTransportURI when stopping
This only describes the transport of the file across the network, not the synergistic relationship of your Audirvāna configuration(s) to the OS, etc… Maybe not relevant, but maybe there is something lurking there in that synergy.
